Hello WOrld?

Selecting Best Performing Model

The table compares several different configurations of an LGBM model (based on tuning metric and included features).

	Model	auc	pr_auc	_f1_micro	_f1_macro	logloss	accuracy	precision_macro	recall_macro	f1_macro	target_f1	target_recall	target_precision	fbeta_1.5	fbeta_2.5	fbeta_4.0	log_loss	elapsed_time	total_size
0	LGBM_Dart_AUC	0.79	0.28	0.73	0.56	9.74	0.73	0.58	0.72	0.56	0.30	0.70	0.19	0.38	0.51	0.61	0.54	204.60	21.87
1	LGBM_AUC	0.79	0.28	0.74	0.57	9.54	0.74	0.58	0.72	0.57	0.30	0.69	0.19	0.38	0.51	0.60	0.53	73.70	23.59
2	LGBM_AUC_All_Features	0.79	0.28	0.74	0.57	9.42	0.74	0.58	0.71	0.57	0.30	0.68	0.19	0.38	0.50	0.59	0.53	57.00	39.72
3	LGBM_AUC_Base_Features	0.77	0.25	0.71	0.55	10.33	0.71	0.57	0.70	0.55	0.28	0.67	0.17	0.36	0.48	0.57	0.56	20.70	21.20
4	LGBM_Weighted_LogLoss	0.76	0.24	0.73	0.56	9.61	0.73	0.57	0.69	0.56	0.28	0.64	0.18	0.36	0.47	0.56	0.54	18.80	11.85
5	Baseline_Only_CreditRatings	0.73	0.20	0.68	0.52	11.63	0.68	0.55	0.67	0.52	0.25	0.65	0.15	0.32	0.45	0.55	0.61	2.30	2.14

We have selected LGBM_Dart_AUC as our final "production" model. LightGBM be default issues 'GBDT' (Gradient Boosting Decision Tree) as its boosting algorithm. 'DART' (Dropouts meet Multiple Additive Regression Trees ) is an alternativeboosting algorithm variant that helps prevent overfitting and improve model generalization by randomly dropping a fraction of boosting trees during training (similar to dropout in neural networks).

Because DART's process of dropping and adding trees back into the model requires more computational work and longer training periods, additionally the stochastic nature of the dropout process can lead to less consistent performance across different training runs compared to GBDT which might be problematic when tuning.

However, we have chose to use LGBM + DART as our primary model because it provides slightly better performance and it's theoretical advantages (reduce overfitting and better Generalization) outweigh the slower training time (5 fold CV + fitting a model on the full train set only take ~2 minutes).

V:\projects\ppuodz-ML.4.1\shared\graph.py:1057: UserWarning: The figure layout has changed to tight
  plt.tight_layout()

Overal we can see that while we are able relatively high AUC of 0.78-0.79 the classification performance, is still very poor. Specifically the very large rate of false positives is a concern because it would mean that ~32% of loans in the sample which were not defaulted on would be rejected when using our model.

On the other hand a somewhat large proportion of (~65%) of problematic loans were detected as such, considering the inherent risks of catering to clients with poorly know credit history that's a relative reasonable result.

We'll investigate this further but generally we can see that the model might be useful to Home Credit if they decided to take a more conservative approach to granting loans.

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x0000029A7FC24CC0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x0000029A7FC24400> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x0000029A7FC242C0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)

Appending previous history
drop drop_cols_post_proc: 227
after drop_cols_post_proc: 121
Full DS size: 307511

array([0.52224028, 0.38090953, 0.60392955, ..., 0.77732527, 0.17447063,
       0.30083482])

101626    0.522240
155435    0.380910
14857     0.603930
123290    0.265638
32502     0.094382
            ...   
292358    0.455475
43225     0.749569
213138    0.777325
138631    0.174471
233831    0.300835
Name: TARGET, Length: 23064, dtype: float64

pandas.core.series.Series

Feature Importance

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\shap\explainers\_tree.py:448: UserWarning: LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray
  warnings.warn('LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray')

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\shap\explainers\_tree.py:448: UserWarning: LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray
  warnings.warn('LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray')

Calculating interest rates:

To determine interest rates for each loan while accounting for default risk and aiming to maximize total return on a portfolio, we can employ several advanced modeling techniques. Determening appropriate interest rates gennerally involves not only predicting the risk of default but also integrating this risk assessment into pricing strategies that reflect the level of risk associated with each loan.

We'll use a simple Risk-Based Pricing Modelin our calculation: it directly links the interest rate charged on a loan to the estimated risk of default. The basic steps to build such a model are:

• Default Probability Estimation: We'll use the probability of default for each loan.
• Loss Given Default (LGD) and Expected Loss: is the amount of loss if the borrower defaults. This can be a fixed percentage or varied based on collateral or loan type (i.e. usually . The Expected Loss (EL) is then calculated as EL = PD × LGD
• where PD is the probability of default.
• We'll set the interest rate such that it covers the expected loss, plus a profit margin. For instance, if the cost of funds is 3%, the EL is 2%, and the desired profit margin is 2%, the minimum interest rate might be set at 7%.

Additionally more complex models could (and probably should) be used in real world scenarios that would incorporate the market conditions (i.e. base interbank rates, margins offered by competitors etc.).

2. Profit Maximization Model This model focuses on maximizing the total return by considering both the interest income and the potential losses due to defaults.

Optimization Framework: Use an optimization model that calculates the optimal interest rate for each loan type. This model would use inputs from the risk model (PD and LGD) and incorporate constraints like minimum return requirements, risk appetite, and regulatory requirements. Simulation Techniques: Simulate different interest rate scenarios for various risk levels to determine the interest rate that maximizes profit while keeping the default risk within acceptable bounds.

3. Portfolio Approach Instead of setting rates on a loan-by-loan basis, consider the entire portfolio's risk and return.

Portfolio Diversification: Assess the risk contribution of each loan type to the overall portfolio and adjust interest rates to achieve desired diversification and risk-return profile. Risk-adjusted Return on Capital (RAROC): Utilize RAROC to evaluate the profitability of a loan, considering the capital at risk. RAROC is used to ensure that the adjusted return (considering risk) meets a certain threshold.

Calibrating Probabilities

Mean. Model Predicted Probabilities = 39.09%

Actual Portfolio Default Rate = 8.07%

We can't directly use the probabilities output by our model to perform any financial analysis because they are not aligned with the true probabilities, which is necessary for risk-based pricing models.

We'll have to calibrate our probabilities using CalibratedClassifierCV and using Isotonic Regression (A non-parametric approach that provides a piecewise linear calibration)

0.29301075268817206

Mean. Model Calibrated Probabilities = 8.17%

Calculating hypothetical interest rates

The application_test.csv does not contain data on the interest rate for granted loans which makes it impossible to calculate the "base" rate used by Home Credit. For our example we'll use a base interest rate of 4% (round 12 month Euribor) and a margin of 6%.

We'll set the LGD to a constant value of 0.5. Generally LGD often ranges between 20% to 60% depending on the type of loan and collateral. For unsecured loans, LGD tends to be higher (closer to or exceeding 50%) due to the lack of recoverable assets.

Calibrated Probabilities mean:  8.17%
Loss Given Default (LGD):  0.5
Expected Loss (EL):  0.04083048044259347
Interest Rates: 14.083%

Asigning Loan Grades

[{'label': 'A Grade', 'start': 0.0, 'end': 0.17682922844255478, 'color': 'grey', 'description': 'Default rate up to 1%'}, {'label': 'B Grade', 'start': 0.17682922844255478, 'end': 0.36787168375037776, 'color': (0.996078431372549, 0.8497808535178777, 0.46145328719723183), 'description': 'Default rate up to 3%'}, {'label': 'C Grade', 'start': 0.36787168375037776, 'end': 0.5064617183285625, 'color': (0.9920953479430988, 0.5490657439446367, 0.23418685121107266), 'description': 'Default rate up to 7%'}, {'label': 'D Grade', 'start': 0.5064617183285625, 'end': 0.7781111185306839, 'color': (0.8866897347174163, 0.09956170703575548, 0.11072664359861592), 'description': 'Default rate up to 15%'}, {'label': 'E-F-G Grade', 'start': 0.7781111185306839, 'end': 1.0, 'color': 'red', 'description': 'Default rate up to 100%'}]

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))

The chart shows the performance of the if only individual with stroke Prob. > T are selected. Additionally the overlay indicates the number of people whose predicted P is in an given range. The overlays can be used to selected the most at risk individual based on the probability predicted for them

C:\Users\Paulius\AppData\Local\Temp\ipykernel_24596\4067243830.py:35: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
  .apply(local_utils.calculate_metrics)

'Returns and Default Rate Based on Default Probability Model:'

	default_rate	mean_interest_rate	expected_return
predicted_grades
A	0.010	0.104	0.103
B	0.030	0.114	0.111
C	0.070	0.134	0.125
D	0.150	0.176	0.150
E-F-G	0.351	0.289	0.188

0.00711064862989941

Loan Acception Criteria and Profit Calculation

The final step we can attempt to is to try to calculate the hypothetical real return of Home Credit's loan portfolio if they used our model to employ a more conservative lending strategy (i.e. rejected all loans where the pre-calibrated default probability is > 0.5).

Again this is only an example because we are still using the interest rates calculated using the calibrated default probabilities should it should be replaced with actual interest rates from Home Credit's loan's to make the data actually meaningful.

	Actual	Hypothetical
Total Loan Amount	13921.5362M	10238.6082M
Total Interest Paid	18.1365M	14.4314M
Total Return %	0.13%	0.14%
Default Rate	7.97%	3.32%
Total Loss	1025.7732M	167.8826M
Losses Avoided	None	857.8905M
Interest Lost	None	-3.7051M
Total Applications Accepted	23064	16306

That being said Total Loss and Losses Avoided figures are likely to be more accurate (we're using an static LGD of 60% which again should be replaced with the real value from Home Credit's data) than the total return. We can see that we could effectively reduce the total losses by ~80% by ~30% less applications.

Explainability

We've used LGBM which is relatively complex "blackbox" model which might not be the ideal in loan evaluations and similar tasks because it's hard to objectively explain the specific decisions the model made (based on regulatory or customer related requirements).

However, we believe that we were largely able to overcome this shortcoming through the use of single observation SHAP plots:

they allow us to attribute the impact of specific feature (e.g. credit scores, client income etc.) on the estimated risk which allows to select an appropriate grade, interest rate and decide whether the loan should or should not be approved based on our acceptable risk preferences:

Each plotted line explains a single model prediction.

The chart above visualizes the model predictions process for individual observations (for a subsample of 100 loans). For a single sample, the line charts the path from the base value to the final predicted value. ach feature causes the line to shift up or down. This shift is determined by its SHAP value. indicates a feature pushing the prediction higher, while blue indicates it pushes the prediction lower.

Individual Predictions

Below we have included some individuals predictions (focusing a the least accurate, most accurate and some random cases in between). In theory these plots can be used as "continious" decision tree and used to explain why a specific loan was accepted or not for regulatory or other purposes.

e.g. Actual Y Value: 1, Predicted Propability: 0.96 indicates that the application had payment difficulties and that our model predicted this outcome with a 96% likelihood.

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\shap\explainers\_tree.py:448: UserWarning: LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray
  warnings.warn('LightGBM binary classifier with TreeExplainer shap values output has changed to a list of ndarray')

Appendix

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_scorer.py:548: FutureWarning: The `needs_threshold` and `needs_proba` parameter are deprecated in version 1.4 and will be removed in 1.6. You can either let `response_method` be `None` or set it to `predict` to preserve the same behaviour.
  warnings.warn(

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x0000019C5C314400> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x0000019C5C314CC0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x0000019C5C3142C0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)

Appending previous history
Full DS size: 307511

	Total NaN Values	Proportion NaN (%)
PrevRatioRejectedAccepted	16847	5.0

Exploratory Analysis

This notebooks includes the analysis of selected variables (based on their importance at predicting the target variable) and their relationships. Individual analysis of each variable is available in the EDA_appendices notebook.

	Total NaN Values	Proportion NaN (%)
ExtSource2	660	0.0
ExtSource3	60965	20.0
ExtSource1	173378	56.0
AmtGoodsPrice	278	0.0
OwnCarAge	202929	66.0
PrevAmtDownPaymentSum	16454	5.0
AmtAnnuity	12	0.0
MeanbureaudaysCredit	44020	14.0
MeanbureauamtCreditSumDebt	51380	17.0
PrevAvgYieldGroup	18945	6.0
PrevCreditReceivedRequestedDiff	16454	5.0
OccupationType	96391	31.0
PrevRatioRejectedAccepted	16847	5.0
MaxbureaudaysCreditEnddate	46269	15.0
PrevLastLoanGoodsCategory	16454	5.0
MeanbureauamtCreditMaxOverdue	123625	40.0

V:\projects\ppuodz-ML.4.1\shared\graph.py:1269: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  corr = round(corr.applymap(pd.to_numeric), 2)

The TARGET variable (loans with payment difficulties) is most correlated with credit ratings obtained from external sources. The correlation is very weak but still significant.

	Coefficient	P-Value
Column
ExtSource3	-0.161	0.000
ExtSource1	-0.131	0.000
ExtSource2	-0.128	0.000
MeanbureaudaysCredit	0.093	0.000
OccupationType	0.075	0.000
DaysEmployed	0.074	0.000
PrevRatioRejectedAccepted	0.073	0.000
PrevRatioRejectedAccepted_cats	0.072	0.000
PrevRatioRejectedAccepted_cats_2	0.072	0.000
OrganizationType	0.069	0.000
NameEducationType	0.067	0.000
PrevAmtDownPaymentSum	-0.057	0.000
PrevCreditReceivedRequestedDiff	0.055	0.000
DaysBirth	0.053	0.000
PrevLastLoanGoodsCategory	0.051	0.000
OwnCarAge	0.050	0.000
MeanbureauamtCreditSumDebt	0.049	0.000
MeanbureauamtCreditMaxOverdue	0.044	0.000
DaysIdPublish	0.042	0.000
CodeGender	0.041	0.000
PrevAvgYieldGroup	0.040	0.000
FlagDocument3	0.039	0.000
AmtGoodsPrice	-0.034	0.000
MaxbureaudaysCreditEnddate	0.034	0.000
NameFamilyStatus	0.027	0.002
AmtCredit	-0.023	0.001

`` Because the datatypes of features vary we had to use different methods to measure the strength and significance of each pair:

• Chi-Squared Test: Assesses independence between two categorical variables. For bool-bool pairs due to categorical nature.

• Point Biserial Correlation: Measures correlation between a binary and a continuous variable. For bool-numerical pairs to account for mixed data types.

• Spearman's Rank Correlation: Assesses monotonic relationship between two continuous variables. Used for numerical-numerical pairs (for non-normally distributed data).

Since the Chi-Squared test outputs an unbound statistic/value which can't be directly compared to pointbiserialr or Spearman Rank we have converted them to a Cramér's V: value which is normalized between 0 and 1. This was done to make the values in the matrix more uniform however we must note that Cramér's V and Spearman's correlation coefficients are fundamentally different statistics and generally can't be directly compared.

CategoricalDtype(categories=['< 25% Rejected', '> 25% Rejected', 'All Accepted', 'No Previous App.'], ordered=False, categories_dtype=object)

The chart below shows the relationship between selected categorical variables and loan status. E.g. a significantly higher proportion of loans taken out by males had issues.

CategoricalDtype(categories=['< 25% Rejected', '> 25% Rejected', 'All Accepted', 'No Previous App.'], ordered=False, categories_dtype=object)

Relationships Between Numerical and Categorical Variables

The charts below show pairs of numerical and categorical features (including some binned numerical features) that have a signficant relationships and at least a small effect size (eta_squared>0.01) based on the non-parametric Kruskal-Wallis Test (one-way ANOVA on ranks) testing whether samples originate from the same distribution.

*It's similar to the Mann–Whitney U test but allows comparing more than 2 groups

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1470: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1483: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

External Credit Scores (ExtSource1)

C:\Users\Paulius\AppData\Local\Temp\ipykernel_35252\2151574185.py:16: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed two minor releases later. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap(obj)`` instead.
  colors = plt.cm.get_cmap('tab10', 4)

	Coefficient	Standard Error	P-Value	Conf. Interval Lower	Conf. Interval Upper
const	0.600	0.040	0.0	0.521	0.680
ExtSource1	-2.099	0.061	0.0	-2.219	-1.979
ExtSource2	-1.964	0.060	0.0	-2.082	-1.846
ExtSource3	-2.779	0.062	0.0	-2.902	-2.657

Normalized credit ratings from three sources are inversely related to default risk, with ExtSource3 having the strongest influence. We can see that a basic Logistic model can already provide a reasonably high result (AUC = 0.74). However, we have to note that the results are based on the full training set and are only provided for EDA/feature analysis purposes. Full statistical modelling will be done in further sections.

C:\Users\Paulius\AppData\Local\Temp\ipykernel_35252\3397180186.py:5: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 1][col], label=f'{col} - Default', shade=True)
C:\Users\Paulius\AppData\Local\Temp\ipykernel_35252\3397180186.py:6: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 0][col], label=f'{col} - No Default', shade=True)

PrevRatioRejectedAccepted_cats
All Accepted        190370
> 25% Rejected       66215
< 25% Rejected       34079
No Previous App.     16847
Name: count, dtype: int64

TotalDefaults_cats
No Defaults          304114
1 Defaulted Loans      3397
Name: count, dtype: int64

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
File ~\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\pandas\core\indexes\base.py:3805, in Index.get_loc(self, key)
   3804 try:
-> 3805     return self._engine.get_loc(casted_key)
   3806 except KeyError as err:

File index.pyx:167, in pandas._libs.index.IndexEngine.get_loc()

File index.pyx:196, in pandas._libs.index.IndexEngine.get_loc()

File pandas\\_libs\\hashtable_class_helper.pxi:7081, in pandas._libs.hashtable.PyObjectHashTable.get_item()

File pandas\\_libs\\hashtable_class_helper.pxi:7089, in pandas._libs.hashtable.PyObjectHashTable.get_item()

KeyError: 'TotalDefaults'

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
Cell In[22], line 1
----> 1 features_matrix_with_bins["TotalDefaults"].value_counts()

File ~\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\pandas\core\frame.py:4102, in DataFrame.__getitem__(self, key)
   4100 if self.columns.nlevels > 1:
   4101     return self._getitem_multilevel(key)
-> 4102 indexer = self.columns.get_loc(key)
   4103 if is_integer(indexer):
   4104     indexer = [indexer]

File ~\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\pandas\core\indexes\base.py:3812, in Index.get_loc(self, key)
   3807     if isinstance(casted_key, slice) or (
   3808         isinstance(casted_key, abc.Iterable)
   3809         and any(isinstance(x, slice) for x in casted_key)
   3810     ):
   3811         raise InvalidIndexError(key)
-> 3812     raise KeyError(key) from err
   3813 except TypeError:
   3814     # If we have a listlike key, _check_indexing_error will raise
   3815     #  InvalidIndexError. Otherwise we fall through and re-raise
   3816     #  the TypeError.
   3817     self._check_indexing_error(key)

KeyError: 'TotalDefaults'

Previous Loan History and Default Risk

The chart below shows the default rate based on whether applicant has previous applied for loans with Home Cred: No Previous App. - no previous applications for client found (i.e. new clients) All Accepted - all previous applications were accepted < 25% Rejected - less than 1/4 applications were rejected > 25% Rejected - more than 1/4 applications were rejected

Interestingly we can see that while applicants whose previous loans were rejected are significantly more likely to default when finally given a loan previous clients with no failed applications have a higher default risk than new clients.

This likely limits the usefulness of the previous_application table because only a small proportion of clients have previously rejected applications

Full DS size: 307511

'Distribution of Samples'

dict_keys(['LGBM_Dart_AUC_NEW'])

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x000001D77E574400> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x000001D77E5742C0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x000001D77E574CC0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)

Appending previous history

[I 2024-04-30 14:10:40,243] A new study created in memory with name: no-name-4805c4d6-1ec9-4777-906c-84869c0c80c2

drop drop_cols_post_proc: 227
after drop_cols_post_proc: 121
Full DS size: 307511
Tunning: - transformers: transformers: 0
 total options: 0
[]
search_grid:
{}

 - model: LGBM_Dart_AUC_NEW n_iters=10 with:
 {'model__class_weight': ['balanced', None], 'model__objective': ['binary'], 'model__boosting_type': ['gbdt', 'rf', 'dart'], 'model__n_estimators': Range(50, 1000, 50, int), 'model__learning_rate': Range(0.01, 0.3, 0.01, float), 'model__max_depth': Range(3, 11, 1, int), 'model__num_leaves': Range(8, 256, 8, int), 'model__min_gain_to_split': Range(0.0, 15.0, 0.5, float), 'model__min_data_in_leaf': Range(0, 3000, 100, int), 'model__lambda_l1': Range(0, 110, 5, int), 'model__lambda_l2': Range(0, 110, 5, int), 'model__bagging_fraction': Range(0.2, 1.0, 0.1, float), 'model__feature_fraction': Range(0.2, 1.0, 0.1, float), 'model__max_bin': Range(50, 500, 25, int), 'model__drop_rate': Range(0.0, 1.0, 0.025, float)}
Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7248, std_test_score:0.00209 train_set_score:0.7308
folds val/train: [0.727, 0.7253, 0.7214, 0.7266, 0.7235] / [0.7308, 0.7311, 0.732, 0.7298, 0.7304], mean fold time: 1.24

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
[I 2024-04-30 14:10:47,163] Trial 0 finished with value: 0.7247591221654156 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'rf', 'model__n_estimators': 900, 'model__learning_rate': 0.08, 'model__max_depth': 11, 'model__num_leaves': 112, 'model__min_gain_to_split': 12.5, 'model__min_data_in_leaf': 2700, 'model__lambda_l1': 50, 'model__lambda_l2': 15, 'model__bagging_fraction': 0.4, 'model__feature_fraction': 0.8, 'model__max_bin': 225, 'model__drop_rate': 0.55}. Best is trial 0 with value: 0.7247591221654156.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7332, std_test_score:0.00291 train_set_score:0.7374
folds val/train: [0.7347, 0.732, 0.729, 0.7378, 0.7323] / [0.7368, 0.7383, 0.7361, 0.7382, 0.7374], mean fold time: 5.41

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
[I 2024-04-30 14:11:14,958] Trial 1 finished with value: 0.733164335671028 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 500, 'model__learning_rate': 0.01, 'model__max_depth': 4, 'model__num_leaves': 216, 'model__min_gain_to_split': 7.0, 'model__min_data_in_leaf': 1400, 'model__lambda_l1': 60, 'model__lambda_l2': 75, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 200, 'model__drop_rate': 0.05}. Best is trial 1 with value: 0.733164335671028.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7266, std_test_score:0.00277 train_set_score:0.7356
folds val/train: [0.7285, 0.7245, 0.7238, 0.7311, 0.7252] / [0.7353, 0.7356, 0.7359, 0.7363, 0.7348], mean fold time: 1.25

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
[I 2024-04-30 14:11:21,928] Trial 2 finished with value: 0.726644314510903 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'rf', 'model__n_estimators': 750, 'model__learning_rate': 0.2, 'model__max_depth': 7, 'model__num_leaves': 160, 'model__min_gain_to_split': 10.5, 'model__min_data_in_leaf': 1500, 'model__lambda_l1': 30, 'model__lambda_l2': 20, 'model__bagging_fraction': 0.9000000000000001, 'model__feature_fraction': 0.2, 'model__max_bin': 175, 'model__drop_rate': 0.30000000000000004}. Best is trial 1 with value: 0.733164335671028.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7681, std_test_score:0.00153 train_set_score:0.8038
folds val/train: [0.7694, 0.7654, 0.7673, 0.7693, 0.769] / [0.8039, 0.8033, 0.8042, 0.8038, 0.8038], mean fold time: 21.81

[I 2024-04-30 14:13:11,758] Trial 3 finished with value: 0.7680802042485764 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 350, 'model__learning_rate': 0.08, 'model__max_depth': 9, 'model__num_leaves': 144, 'model__min_gain_to_split': 7.5, 'model__min_data_in_leaf': 700, 'model__lambda_l1': 90, 'model__lambda_l2': 0, 'model__bagging_fraction': 0.6000000000000001, 'model__feature_fraction': 0.4, 'model__max_bin': 375, 'model__drop_rate': 0.6000000000000001}. Best is trial 3 with value: 0.7680802042485764.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7713, std_test_score:0.00130 train_set_score:0.8122
folds val/train: [0.7734, 0.7696, 0.7704, 0.771, 0.7718] / [0.8116, 0.8118, 0.8124, 0.813, 0.8122], mean fold time: 2.36

[I 2024-04-30 14:13:24,340] Trial 4 finished with value: 0.771269722930656 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 950, 'model__learning_rate': 0.11, 'model__max_depth': 6, 'model__num_leaves': 224, 'model__min_gain_to_split': 13.0, 'model__min_data_in_leaf': 0, 'model__lambda_l1': 15, 'model__lambda_l2': 80, 'model__bagging_fraction': 0.9000000000000001, 'model__feature_fraction': 0.4, 'model__max_bin': 500, 'model__drop_rate': 0.875}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7296, std_test_score:0.00302 train_set_score:0.7395
folds val/train: [0.7331, 0.7296, 0.7247, 0.7322, 0.7283] / [0.7379, 0.7399, 0.7407, 0.7397, 0.7391], mean fold time: 1.35

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\sklearn\metrics\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, f"{metric.capitalize()} is", len(result))
[I 2024-04-30 14:13:31,840] Trial 5 finished with value: 0.7295734635518236 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'rf', 'model__n_estimators': 600, 'model__learning_rate': 0.17, 'model__max_depth': 11, 'model__num_leaves': 256, 'model__min_gain_to_split': 4.0, 'model__min_data_in_leaf': 300, 'model__lambda_l1': 110, 'model__lambda_l2': 85, 'model__bagging_fraction': 1.0, 'model__feature_fraction': 0.7, 'model__max_bin': 275, 'model__drop_rate': 0.325}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7702, std_test_score:0.00128 train_set_score:0.8221
folds val/train: [0.7725, 0.7697, 0.7689, 0.7692, 0.7706] / [0.8388, 0.8306, 0.8252, 0.8057, 0.81], mean fold time: 2.27

[I 2024-04-30 14:13:44,003] Trial 6 finished with value: 0.7701777822813336 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 450, 'model__learning_rate': 0.24000000000000002, 'model__max_depth': 8, 'model__num_leaves': 24, 'model__min_gain_to_split': 4.5, 'model__min_data_in_leaf': 2800, 'model__lambda_l1': 0, 'model__lambda_l2': 70, 'model__bagging_fraction': 0.30000000000000004, 'model__feature_fraction': 1.0, 'model__max_bin': 450, 'model__drop_rate': 0.225}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7646, std_test_score:0.00148 train_set_score:0.7801
folds val/train: [0.765, 0.7623, 0.7635, 0.7657, 0.7664] / [0.7801, 0.7805, 0.7804, 0.78, 0.7794], mean fold time: 3.33

[I 2024-04-30 14:14:01,738] Trial 7 finished with value: 0.7645755916988988 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 850, 'model__learning_rate': 0.11, 'model__max_depth': 8, 'model__num_leaves': 192, 'model__min_gain_to_split': 14.0, 'model__min_data_in_leaf': 2600, 'model__lambda_l1': 25, 'model__lambda_l2': 105, 'model__bagging_fraction': 0.2, 'model__feature_fraction': 0.9000000000000001, 'model__max_bin': 225, 'model__drop_rate': 0.17500000000000002}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7633, std_test_score:0.00208 train_set_score:0.7764
folds val/train: [0.7656, 0.7601, 0.7617, 0.764, 0.765] / [0.7759, 0.7759, 0.7766, 0.777, 0.7768], mean fold time: 1.92

[I 2024-04-30 14:14:12,368] Trial 8 finished with value: 0.7632609111571824 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 550, 'model__learning_rate': 0.28, 'model__max_depth': 5, 'model__num_leaves': 240, 'model__min_gain_to_split': 14.5, 'model__min_data_in_leaf': 0, 'model__lambda_l1': 20, 'model__lambda_l2': 90, 'model__bagging_fraction': 1.0, 'model__feature_fraction': 0.6000000000000001, 'model__max_bin': 100, 'model__drop_rate': 0.9}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7257, std_test_score:0.00200 train_set_score:0.7356
folds val/train: [0.7258, 0.7237, 0.7239, 0.7293, 0.7258] / [0.7357, 0.7354, 0.7359, 0.7356, 0.7351], mean fold time: 2.34

[I 2024-04-30 14:14:25,096] Trial 9 finished with value: 0.7256979936912922 and parameters: {'model__class_weight': None, 'model__objective': 'binary', 'model__boosting_type': 'rf', 'model__n_estimators': 650, 'model__learning_rate': 0.05, 'model__max_depth': 10, 'model__num_leaves': 56, 'model__min_gain_to_split': 7.0, 'model__min_data_in_leaf': 2100, 'model__lambda_l1': 15, 'model__lambda_l2': 5, 'model__bagging_fraction': 0.30000000000000004, 'model__feature_fraction': 0.9000000000000001, 'model__max_bin': 75, 'model__drop_rate': 0.625}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7638, std_test_score:0.00140 train_set_score:0.7937
folds val/train: [0.7656, 0.7613, 0.7638, 0.7642, 0.7643] / [0.7935, 0.7939, 0.794, 0.7938, 0.7933], mean fold time: 1.91

[I 2024-04-30 14:14:35,466] Trial 10 finished with value: 0.7638478396414092 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 50, 'model__learning_rate': 0.12, 'model__max_depth': 6, 'model__num_leaves': 88, 'model__min_gain_to_split': 1.0, 'model__min_data_in_leaf': 700, 'model__lambda_l1': 50, 'model__lambda_l2': 45, 'model__bagging_fraction': 0.7, 'model__feature_fraction': 0.5, 'model__max_bin': 500, 'model__drop_rate': 0.9750000000000001}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7709, std_test_score:0.00211 train_set_score:0.7988
folds val/train: [0.7743, 0.7703, 0.7685, 0.7691, 0.7722] / [0.807, 0.8008, 0.7974, 0.7913, 0.7973], mean fold time: 2.18

[I 2024-04-30 14:14:47,169] Trial 11 finished with value: 0.7708781891661243 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 300, 'model__learning_rate': 0.23, 'model__max_depth': 3, 'model__num_leaves': 8, 'model__min_gain_to_split': 3.5, 'model__min_data_in_leaf': 2100, 'model__lambda_l1': 0, 'model__lambda_l2': 60, 'model__bagging_fraction': 0.8, 'model__feature_fraction': 1.0, 'model__max_bin': 475, 'model__drop_rate': 0.775}. Best is trial 4 with value: 0.771269722930656.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7730, std_test_score:0.00203 train_set_score:0.7982
folds val/train: [0.7767, 0.7722, 0.7707, 0.7722, 0.7731] / [0.8062, 0.797, 0.795, 0.7971, 0.7957], mean fold time: 6.12

[I 2024-04-30 14:15:19,006] Trial 12 finished with value: 0.7729736600713923 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 250, 'model__learning_rate': 0.21000000000000002, 'model__max_depth': 3, 'model__num_leaves': 16, 'model__min_gain_to_split': 0.0, 'model__min_data_in_leaf': 2000, 'model__lambda_l1': 0, 'model__lambda_l2': 50, 'model__bagging_fraction': 0.8, 'model__feature_fraction': 0.5, 'model__max_bin': 375, 'model__drop_rate': 0.8}. Best is trial 12 with value: 0.7729736600713923.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7673, std_test_score:0.00125 train_set_score:0.7823
folds val/train: [0.7684, 0.7661, 0.7661, 0.7666, 0.7692] / [0.782, 0.7827, 0.7823, 0.7819, 0.7825], mean fold time: 3.38

[I 2024-04-30 14:15:36,908] Trial 13 finished with value: 0.7672730632046433 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 100, 'model__learning_rate': 0.16, 'model__max_depth': 3, 'model__num_leaves': 184, 'model__min_gain_to_split': 0.0, 'model__min_data_in_leaf': 2000, 'model__lambda_l1': 0, 'model__lambda_l2': 40, 'model__bagging_fraction': 0.8, 'model__feature_fraction': 0.5, 'model__max_bin': 400, 'model__drop_rate': 0.775}. Best is trial 12 with value: 0.7729736600713923.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7689, std_test_score:0.00186 train_set_score:0.7956
folds val/train: [0.7712, 0.7661, 0.7674, 0.7699, 0.7697] / [0.7959, 0.7955, 0.7953, 0.7962, 0.7952], mean fold time: 2.21

[I 2024-04-30 14:15:48,764] Trial 14 finished with value: 0.7688614622014904 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 1000, 'model__learning_rate': 0.3, 'model__max_depth': 5, 'model__num_leaves': 80, 'model__min_gain_to_split': 10.0, 'model__min_data_in_leaf': 1100, 'model__lambda_l1': 65, 'model__lambda_l2': 35, 'model__bagging_fraction': 0.8, 'model__feature_fraction': 0.4, 'model__max_bin': 350, 'model__drop_rate': 0.8}. Best is trial 12 with value: 0.7729736600713923.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7710, std_test_score:0.00142 train_set_score:0.7985
folds val/train: [0.7735, 0.7698, 0.7705, 0.7697, 0.7717] / [0.7988, 0.7982, 0.7994, 0.7984, 0.7977], mean fold time: 1.99

[I 2024-04-30 14:15:59,500] Trial 15 finished with value: 0.7710315875391158 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 200, 'model__learning_rate': 0.19, 'model__max_depth': 6, 'model__num_leaves': 120, 'model__min_gain_to_split': 10.0, 'model__min_data_in_leaf': 1900, 'model__lambda_l1': 35, 'model__lambda_l2': 60, 'model__bagging_fraction': 0.6000000000000001, 'model__feature_fraction': 0.2, 'model__max_bin': 325, 'model__drop_rate': 1.0}. Best is trial 12 with value: 0.7729736600713923.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7676, std_test_score:0.00100 train_set_score:0.7938
folds val/train: [0.7684, 0.7662, 0.7665, 0.7681, 0.7687] / [0.7934, 0.7942, 0.7944, 0.7936, 0.7934], mean fold time: 38.04

[I 2024-04-30 14:19:10,473] Trial 16 finished with value: 0.7675725766548119 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 300, 'model__learning_rate': 0.13, 'model__max_depth': 4, 'model__num_leaves': 48, 'model__min_gain_to_split': 2.0, 'model__min_data_in_leaf': 1000, 'model__lambda_l1': 10, 'model__lambda_l2': 100, 'model__bagging_fraction': 0.9000000000000001, 'model__feature_fraction': 0.6000000000000001, 'model__max_bin': 425, 'model__drop_rate': 0.42500000000000004}. Best is trial 12 with value: 0.7729736600713923.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7691, std_test_score:0.00187 train_set_score:0.7993
folds val/train: [0.7723, 0.7666, 0.7685, 0.7687, 0.7696] / [0.7991, 0.7992, 0.8001, 0.7993, 0.7987], mean fold time: 1.62

[I 2024-04-30 14:19:19,385] Trial 17 finished with value: 0.7691357518070465 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 750, 'model__learning_rate': 0.25, 'model__max_depth': 6, 'model__num_leaves': 176, 'model__min_gain_to_split': 11.5, 'model__min_data_in_leaf': 2400, 'model__lambda_l1': 40, 'model__lambda_l2': 75, 'model__bagging_fraction': 0.7, 'model__feature_fraction': 0.4, 'model__max_bin': 300, 'model__drop_rate': 0.7000000000000001}. Best is trial 12 with value: 0.7729736600713923.

Fold: Tuning: n_train=246008, eval_set=61503
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7712, std_test_score:0.00147 train_set_score:0.7943
folds val/train: [0.7729, 0.7687, 0.7706, 0.7714, 0.7724] / [0.7932, 0.7941, 0.796, 0.7939, 0.7941], mean fold time: 2.13

[I 2024-04-30 14:19:31,234] Trial 18 finished with value: 0.7712175359581087 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'gbdt', 'model__n_estimators': 200, 'model__learning_rate': 0.19, 'model__max_depth': 4, 'model__num_leaves': 216, 'model__min_gain_to_split': 8.0, 'model__min_data_in_leaf': 1500, 'model__lambda_l1': 75, 'model__lambda_l2': 45, 'model__bagging_fraction': 0.9000000000000001, 'model__feature_fraction': 0.5, 'model__max_bin': 500, 'model__drop_rate': 0.875}. Best is trial 12 with value: 0.7729736600713923.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7732, std_test_score:0.00156 train_set_score:0.8209
folds val/train: [0.7756, 0.7711, 0.7719, 0.7735, 0.7739] / [0.8221, 0.8203, 0.8207, 0.8213, 0.8203], mean fold time: 25.83

[I 2024-04-30 14:21:41,235] Trial 19 finished with value: 0.7732021037088023 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 400, 'model__learning_rate': 0.14, 'model__max_depth': 7, 'model__num_leaves': 88, 'model__min_gain_to_split': 13.5, 'model__min_data_in_leaf': 0, 'model__lambda_l1': 10, 'model__lambda_l2': 90, 'model__bagging_fraction': 0.7, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 425, 'model__drop_rate': 0.45}. Best is trial 19 with value: 0.7732021037088023.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7737, std_test_score:0.00150 train_set_score:0.8402
folds val/train: [0.7753, 0.7722, 0.7717, 0.7744, 0.7751] / [0.8404, 0.8396, 0.8411, 0.8404, 0.8397], mean fold time: 19.83

[I 2024-04-30 14:23:21,211] Trial 20 finished with value: 0.773736619748256 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 400, 'model__learning_rate': 0.22, 'model__max_depth': 8, 'model__num_leaves': 40, 'model__min_gain_to_split': 5.0, 'model__min_data_in_leaf': 3000, 'model__lambda_l1': 10, 'model__lambda_l2': 110, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 425, 'model__drop_rate': 0.45}. Best is trial 20 with value: 0.773736619748256.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7745, std_test_score:0.00123 train_set_score:0.8443
folds val/train: [0.776, 0.7732, 0.7729, 0.7746, 0.7756] / [0.8443, 0.843, 0.844, 0.8456, 0.8444], mean fold time: 19.61

[I 2024-04-30 14:25:00,094] Trial 21 finished with value: 0.7744710826089047 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 450, 'model__learning_rate': 0.23, 'model__max_depth': 8, 'model__num_leaves': 48, 'model__min_gain_to_split': 5.5, 'model__min_data_in_leaf': 2900, 'model__lambda_l1': 10, 'model__lambda_l2': 110, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 400, 'model__drop_rate': 0.42500000000000004}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7732, std_test_score:0.00163 train_set_score:0.8425
folds val/train: [0.7754, 0.7708, 0.7719, 0.7734, 0.7744] / [0.8433, 0.8433, 0.8425, 0.8428, 0.8406], mean fold time: 18.03

[I 2024-04-30 14:26:31,031] Trial 22 finished with value: 0.7731975885382104 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 400, 'model__learning_rate': 0.25, 'model__max_depth': 8, 'model__num_leaves': 56, 'model__min_gain_to_split': 5.5, 'model__min_data_in_leaf': 2900, 'model__lambda_l1': 10, 'model__lambda_l2': 110, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 450, 'model__drop_rate': 0.42500000000000004}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7736, std_test_score:0.00140 train_set_score:0.8264
folds val/train: [0.7753, 0.7721, 0.7719, 0.7739, 0.7748] / [0.8266, 0.8246, 0.8273, 0.8272, 0.8262], mean fold time: 20.32

[I 2024-04-30 14:28:13,425] Trial 23 finished with value: 0.7736188555317012 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 400, 'model__learning_rate': 0.27, 'model__max_depth': 9, 'model__num_leaves': 80, 'model__min_gain_to_split': 5.5, 'model__min_data_in_leaf': 2400, 'model__lambda_l1': 40, 'model__lambda_l2': 95, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 425, 'model__drop_rate': 0.45}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7743, std_test_score:0.00114 train_set_score:0.8227
folds val/train: [0.7754, 0.7741, 0.7725, 0.7737, 0.7756] / [0.8235, 0.8217, 0.8219, 0.8236, 0.8225], mean fold time: 24.82

[I 2024-04-30 14:30:18,349] Trial 24 finished with value: 0.7742569892640564 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 500, 'model__learning_rate': 0.28, 'model__max_depth': 9, 'model__num_leaves': 40, 'model__min_gain_to_split': 5.5, 'model__min_data_in_leaf': 3000, 'model__lambda_l1': 40, 'model__lambda_l2': 100, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 350, 'model__drop_rate': 0.35000000000000003}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7730, std_test_score:0.00176 train_set_score:0.8582
folds val/train: [0.776, 0.7708, 0.772, 0.7735, 0.7727] / [0.8573, 0.8581, 0.859, 0.8597, 0.8572], mean fold time: 35.85

[I 2024-04-30 14:33:18,399] Trial 25 finished with value: 0.772998044701975 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 700, 'model__learning_rate': 0.3, 'model__max_depth': 9, 'model__num_leaves': 40, 'model__min_gain_to_split': 2.5, 'model__min_data_in_leaf': 3000, 'model__lambda_l1': 25, 'model__lambda_l2': 110, 'model__bagging_fraction': 0.4, 'model__feature_fraction': 0.2, 'model__max_bin': 375, 'model__drop_rate': 0.325}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7740, std_test_score:0.00162 train_set_score:0.8036
folds val/train: [0.7761, 0.7716, 0.7726, 0.7746, 0.7748] / [0.8038, 0.8028, 0.804, 0.8039, 0.8036], mean fold time: 19.51

[I 2024-04-30 14:34:56,749] Trial 26 finished with value: 0.7739512652732408 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 500, 'model__learning_rate': 0.22, 'model__max_depth': 10, 'model__num_leaves': 32, 'model__min_gain_to_split': 8.5, 'model__min_data_in_leaf': 2500, 'model__lambda_l1': 45, 'model__lambda_l2': 100, 'model__bagging_fraction': 0.4, 'model__feature_fraction': 0.2, 'model__max_bin': 325, 'model__drop_rate': 0.15000000000000002}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7738, std_test_score:0.00148 train_set_score:0.7986
folds val/train: [0.7757, 0.772, 0.7726, 0.7734, 0.7754] / [0.7985, 0.7979, 0.7994, 0.7984, 0.7986], mean fold time: 5.78

[I 2024-04-30 14:35:26,433] Trial 27 finished with value: 0.7738168190120491 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 550, 'model__learning_rate': 0.27, 'model__max_depth': 10, 'model__num_leaves': 64, 'model__min_gain_to_split': 8.0, 'model__min_data_in_leaf': 2500, 'model__lambda_l1': 75, 'model__lambda_l2': 100, 'model__bagging_fraction': 0.4, 'model__feature_fraction': 0.2, 'model__max_bin': 300, 'model__drop_rate': 0.025}. Best is trial 21 with value: 0.7744710826089047.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7751, std_test_score:0.00160 train_set_score:0.8100
folds val/train: [0.7774, 0.7736, 0.7732, 0.7748, 0.7763] / [0.8103, 0.8095, 0.81, 0.81, 0.8101], mean fold time: 21.19

[I 2024-04-30 14:37:13,206] Trial 28 finished with value: 0.77509084497967 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 600, 'model__learning_rate': 0.26, 'model__max_depth': 10, 'model__num_leaves': 24, 'model__min_gain_to_split': 6.0, 'model__min_data_in_leaf': 2300, 'model__lambda_l1': 45, 'model__lambda_l2': 95, 'model__bagging_fraction': 0.30000000000000004, 'model__feature_fraction': 0.2, 'model__max_bin': 325, 'model__drop_rate': 0.125}. Best is trial 28 with value: 0.77509084497967.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502
Tune: val_score:0.7750, std_test_score:0.00135 train_set_score:0.8244
folds val/train: [0.7767, 0.7738, 0.7732, 0.7752, 0.7762] / [0.824, 0.8238, 0.8252, 0.8246, 0.8244], mean fold time: 33.81

[I 2024-04-30 14:40:03,043] Trial 29 finished with value: 0.7750172313526884 and parameters: {'model__class_weight': 'balanced', 'model__objective': 'binary', 'model__boosting_type': 'dart', 'model__n_estimators': 850, 'model__learning_rate': 0.28, 'model__max_depth': 11, 'model__num_leaves': 112, 'model__min_gain_to_split': 6.0, 'model__min_data_in_leaf': 2700, 'model__lambda_l1': 55, 'model__lambda_l2': 65, 'model__bagging_fraction': 0.2, 'model__feature_fraction': 0.7, 'model__max_bin': 250, 'model__drop_rate': 0.125}. Best is trial 28 with value: 0.77509084497967.
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246008, eval_set=61503

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

Fold: Tuning: n_train=246009, eval_set=61502

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

<module 'Draft.feature_builder_v2' from 'V:\\projects\\ppuodz-ML.4.1\\Draft\\feature_builder_v2.py'>

Full DS size: 307511



 -- 
 TEST SIZE: 23064 




Baseline_Only_CreditRatings: 2.3 seconds
Full DS size: 307511



 -- 
 TEST SIZE: 23064 




LGBM_AUC_Base_Features: 20.7 seconds
drop drop_cols_post_proc: 121
after drop_cols_post_proc: 59
Full DS size: 307511



 -- 
 TEST SIZE: 23064 




LGBM_Weighted_LogLoss: 18.8 seconds

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x00000207FE1545E0> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x00000207FE154EA0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x00000207FE1544A0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)

Appending previous history
drop drop_cols_post_proc: 227
after drop_cols_post_proc: 121
Full DS size: 307511



 -- 
 TEST SIZE: 23064 




LGBM_AUC: 73.7 seconds

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x00000207FE1545E0> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x00000207FE154EA0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x00000207FE1544A0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)

Appending previous history
drop drop_cols_post_proc: 227
after drop_cols_post_proc: 121
Full DS size: 307511



 -- 
 TEST SIZE: 23064 

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\lightgbm\callback.py:325: UserWarning: Early stopping is not available in dart mode
  _log_warning('Early stopping is not available in dart mode')

LGBM_Dart_AUC: 204.6 seconds

C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\entityset\entityset.py:1914: UserWarning: index SK_BUREAU_ID not found in dataframe, creating new integer column
  warnings.warn(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\woodwork\type_sys\utils.py:40: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.
  pd.to_datetime(
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function min at 0x00000207FE1545E0> is currently using SeriesGroupBy.min. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "min" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function mean at 0x00000207FE154EA0> is currently using SeriesGroupBy.mean. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "mean" instead.
  ).agg(to_agg)
C:\Users\Paulius\AppData\Local\pypoetry\Cache\virtualenvs\ppuodz-ml-4-1-dqELbViF-py3.12\Lib\site-packages\featuretools\computational_backends\feature_set_calculator.py:828: FutureWarning: The provided callable <function max at 0x00000207FE1544A0> is currently using SeriesGroupBy.max. In a future version of pandas, the provided callable will be used directly. To keep current behavior pass the string "max" instead.
  ).agg(to_agg)

Appending previous history
Full DS size: 307511



 -- 
 TEST SIZE: 23064 




LGBM_AUC_All_Features: 57.0 seconds

Baseline_Only_CreditRatings:
  {'model__n_estimators': 300, 'model__learning_rate': 0.25, 'model__max_depth': 3, 'model__num_leaves': 224, 'model__min_gain_to_split': 1.5, 'model__min_data_in_leaf': 100, 'model__lambda_l1': 0, 'model__lambda_l2': 50, 'model__bagging_fraction': 1.0, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 125}
LGBM_AUC_Base_Features:
  {'model__n_estimators': 800, 'model__learning_rate': 0.060000000000000005, 'model__max_depth': 7, 'model__num_leaves': 48, 'model__min_gain_to_split': 2.5, 'model__min_data_in_leaf': 300, 'model__lambda_l1': 80, 'model__lambda_l2': 5, 'model__bagging_fraction': 0.8, 'model__feature_fraction': 0.8, 'model__max_bin': 350}
LGBM_Weighted_LogLoss:
  {'model__n_estimators': 800, 'model__learning_rate': 0.06999999999999999, 'model__max_depth': 10, 'model__num_leaves': 224, 'model__min_gain_to_split': 3.5, 'model__min_data_in_leaf': 100, 'model__lambda_l1': 5, 'model__lambda_l2': 25, 'model__bagging_fraction': 0.4, 'model__feature_fraction': 0.6000000000000001, 'model__max_bin': 50}
LGBM_AUC:
  {'model__n_estimators': 700, 'model__learning_rate': 0.04, 'model__max_depth': 9, 'model__num_leaves': 120, 'model__min_gain_to_split': 0.5, 'model__min_data_in_leaf': 400, 'model__lambda_l1': 105, 'model__lambda_l2': 5, 'model__bagging_fraction': 0.30000000000000004, 'model__feature_fraction': 0.30000000000000004, 'model__max_bin': 500}
LGBM_Dart_AUC:
  {'model__n_estimators': 900, 'model__learning_rate': 0.27, 'model__max_depth': 4, 'model__num_leaves': 160, 'model__min_gain_to_split': 1.5, 'model__min_data_in_leaf': 2200, 'model__lambda_l1': 85, 'model__lambda_l2': 90, 'model__bagging_fraction': 0.5, 'model__feature_fraction': 1.0, 'model__max_bin': 250, 'model__drop_rate': 0.2}
LGBM_AUC_All_Features:
  {'model__n_estimators': 700, 'model__learning_rate': 0.05, 'model__max_depth': 7, 'model__num_leaves': 224, 'model__min_gain_to_split': 4.0, 'model__min_data_in_leaf': 300, 'model__lambda_l1': 40, 'model__lambda_l2': 10, 'model__bagging_fraction': 0.7, 'model__feature_fraction': 0.8, 'model__max_bin': 100}

Model Results Comparison (5-fold CV)

	Model	auc	pr_auc	_f1_micro	_f1_macro	logloss	accuracy	precision_macro	recall_macro	f1_macro	target_f1	target_recall	target_precision	fbeta_1.5	fbeta_2.5	fbeta_4.0	log_loss	elapsed_time	total_size
4	LGBM_Dart_AUC	0.775	0.263	0.731	0.561	9.711	0.731	0.573	0.705	0.561	0.288	0.674	0.183	0.369	0.492	0.582	0.540	204.6	21.866262
3	LGBM_AUC	0.773	0.261	0.738	0.564	9.448	0.738	0.573	0.702	0.564	0.289	0.660	0.185	0.369	0.487	0.573	0.531	73.7	23.586256
5	LGBM_AUC_All_Features	0.773	0.263	0.741	0.566	9.343	0.741	0.574	0.703	0.566	0.291	0.658	0.187	0.370	0.488	0.573	0.526	57.0	39.724908
1	LGBM_AUC_Base_Features	0.759	0.243	0.715	0.547	10.278	0.715	0.566	0.690	0.547	0.272	0.661	0.171	0.352	0.474	0.566	0.561	20.7	21.202541
2	LGBM_Weighted_LogLoss	0.755	0.239	0.748	0.564	9.073	0.748	0.570	0.684	0.564	0.281	0.608	0.182	0.354	0.460	0.535	0.520	18.8	11.854638
0	Baseline_Only_CreditRatings	0.723	0.203	0.670	0.516	11.904	0.670	0.553	0.665	0.516	0.244	0.660	0.150	0.322	0.449	0.550	0.613	2.3	2.143046

Model Results Comparison (Test Set)

	Model	auc	pr_auc	_f1_micro	_f1_macro	logloss	accuracy	precision_macro	recall_macro	f1_macro	target_f1	target_recall	target_precision	fbeta_1.5	fbeta_2.5	fbeta_4.0	log_loss	elapsed_time	total_size
4	LGBM_Dart_AUC	0.790	0.282	0.730	0.564	9.745	0.730	0.577	0.718	0.564	0.296	0.705	0.188	0.381	0.511	0.607	0.541	204.6	21.866262
3	LGBM_AUC	0.789	0.276	0.735	0.567	9.539	0.735	0.577	0.715	0.567	0.297	0.692	0.189	0.380	0.506	0.598	0.533	73.7	23.586256
5	LGBM_AUC_All_Features	0.787	0.275	0.739	0.568	9.419	0.739	0.577	0.713	0.568	0.296	0.682	0.189	0.379	0.502	0.592	0.528	57.0	39.724908
1	LGBM_AUC_Base_Features	0.768	0.246	0.713	0.548	10.328	0.713	0.567	0.695	0.548	0.275	0.672	0.173	0.356	0.481	0.575	0.562	20.7	21.202541
2	LGBM_Weighted_LogLoss	0.763	0.240	0.733	0.558	9.608	0.733	0.569	0.691	0.558	0.280	0.641	0.179	0.357	0.473	0.557	0.541	18.8	11.854638
0	Baseline_Only_CreditRatings	0.729	0.199	0.677	0.521	11.632	0.677	0.554	0.666	0.521	0.246	0.654	0.152	0.324	0.449	0.547	0.613	2.3	2.143046

V:\projects\ppuodz-ML.4.1\shared\graph.py:1057: UserWarning: The figure layout has changed to tight
  plt.tight_layout()

Full DS size: 307511

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[10], line 11
      8 features_matrix = target_model_config[0].model_pipeline_config.load_data(
      9     loader_function=feature_builder_v2.load_datasets_and_prepare_features)
     10 # 
---> 11 features_all, labels_all = pipeline._get_features_labels(features_matrix)
     12 X_train, X_test, y_train, y_test = pipeline.get_deterministic_train_test_split(
     13     features_all, labels_all
     14 )
     16 X_train = X_train.drop(columns=["TARGET"])

NameError: name 'pipeline' is not defined

The calibration curve, also known as a reliability diagram, is a graphical representation used to evaluate the accuracy of predicted probabilities in classification models. It specifically checks how well the predicted probabilities of a model match the actual outcomes.

The ideal calibration curve is a straight line at a 45-degree angle from the bottom left to the top right of the plot. This line, often called the "line of perfect calibration," indicates that the model's predictions are perfectly calibrated. If your model predicts a class with 70% probability, then 70% of the cases that are predicted as such should indeed belong to that class.

Common Patterns and Their Interpretations: Perfect Calibration:

The points lie on the diagonal line from (0,0) to (1,1). Example: If a model predicts an event with 30% probability, then in the long run, that event occurs about 30% of the time when predicted at this probability. Underconfidence:

The curve lies above the diagonal line. The model's probabilities are lower than the true likelihood of the event. For instance, if events the model predicts to happen 60% of the time actually happen 80% of the time, the model is underconfident. Overconfidence:

The curve lies below the diagonal line. The model predicts higher probabilities than what is true. If a prediction of 80% only happens 60% of the time, the model is overconfident.

Expected Calibration Error (ECE): This measures the average difference between the predicted probabilities and the actual outcomes. Lower values indicate better calibration.

Brier Score: Measures the mean squared difference between the predicted probability and the actual outcome. It is a good measure of the accuracy and calibration of the predictions.

Residual Plots show the difference between observed and predicted probabilities. Helps in checking the assumption of homoscedasticity. Ideally, residuals should be randomly dispersed around the central line, and patterns suggest model inadequacies.

homoscedasticity

Based on the chart we see these potential problems: Probability Estimates are Polarized:

The cup-like pattern at the top suggests the model is very confident (probabilities close to 0 or 1) about certain instances but is incorrect, as these points have higher residuals. The bottom lines being straighter and closer to zero indicate that for a range of predicted probabilities, the residuals are consistently low, which means the model performs well in that range.

Model Overconfidence:

The residuals are larger for predictions near 0 or 1 because the log loss penalizes confident incorrect predictions more harshly than less confident ones. This overconfidence is often characteristic of models that are not well-calibrated and could benefit from probability calibration techniques.

Class Imbalance:

This pattern can sometimes emerge from class imbalance if the model is better at predicting the majority class and frequently mispredicts the minority class with high confidence.

Non-linearity in Feature Space:

The curving pattern could also be a sign that the model is not capturing some non-linear relationships between features and the outcome. This might suggest that feature engineering or a more sophisticated model could be helpful.

Mean Absolute Error (MAE) of Residuals: This is the average of the absolute values of the residuals. It gives an idea of the average magnitude of the prediction errors. Annotation: Indicating MAE on the residual plot can help assess the typical error magnitude in a more intuitive way than just viewing the spread of residuals.

Maximum Residual: The maximum value among the residuals can indicate the worst-case scenario for your predictions. Annotation: Marking the maximum residual can alert users to the worst errors the model could make.

In the finance industry, loan grades (or credit scores) are a crucial part of risk management, helping lenders assess the creditworthiness of borrowers. These grades are typically determined based on various factors, including the borrower's credit history, income stability, debt-to-income ratio, and more. The grades reflect the estimated risk of default, and they directly influence the interest rate offered to the borrower. Commonly, loan grades are categorized from 'A' (lowest risk) to 'G' (highest risk), although the specific categories can vary by institution.

A (Lowest Risk): Below 1% default rate. Borrowers with excellent credit histories and very low risk of default. B: 1% to 3% default rate. Borrowers with good credit histories and low risk of default. C: 3% to 7% default rate. Borrowers with average credit histories and moderate risk of default. D: 7% to 15% default rate. Borrowers with below-average credit histories and higher risk of default. E: 15% to 25% default rate. Borrowers with poor credit histories and very high risk of default. F and G (Highest Risk): Above 25% default rate. Borrowers with very poor credit histories and extremely high risk of default.

Full DS size: 307511

	Total NaN Values	Proportion NaN (%)
PrevRatioRejectedAccepted	16847	5.0

Exploratory Analysis

This notebooks includes the analysis of selected variables (based on their importance at predicting the target variable) and their relationships. Individual analysis of each variable is available in the EDA_appendices notebook.

Dataset Summary

NaN Values by Column:

	Total NaN Values	Proportion NaN (%)
ExtSource2	660	0.0
ExtSource3	60965	20.0
ExtSource1	173378	56.0
AmtGoodsPrice	278	0.0
OwnCarAge	202929	66.0
PrevAmtDownPaymentSum	16454	5.0
AmtAnnuity	12	0.0
MeanbureaudaysCredit	44020	14.0
MeanbureauamtCreditSumDebt	51380	17.0
PrevAvgYieldGroup	18945	6.0
PrevCreditReceivedRequestedDiff	16454	5.0
OccupationType	96391	31.0
PrevRatioRejectedAccepted	16847	5.0
MaxbureaudaysCreditEnddate	46269	15.0
PrevLastLoanGoodsCategory	16454	5.0
MeanbureauamtCreditMaxOverdue	123625	40.0

Duplicates Values

'Duplicated Values: 0'

'Total Columns: 229'

Correlations

Because we has such a large number of columns we have only included features whhich have an importance value { > X } with our final LGBM model: TODO

V:\projects\ppuodz-ML.4.1\shared\graph.py:1276: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  corr = round(corr.applymap(pd.to_numeric), 2)

The TARGET variable (loans with payment difficulties) is most correlated with credit ratings obtained from external sources. The correlation is very weak but still significant.

`` Because the datatypes of features vary we had to use different methods to measure the strength and significance of each pair:

• Chi-Squared Test: Assesses independence between two categorical variables. For bool-bool pairs due to categorical nature.

• Point Biserial Correlation: Measures correlation between a binary and a continuous variable. For bool-numerical pairs to account for mixed data types.

• Spearman's Rank Correlation: Assesses monotonic relationship between two continuous variables. Used for numerical-numerical pairs (for non-normally distributed data).

Since the Chi-Squared test outputs an unbound statistic/value which can't be directly compared to pointbiserialr or Spearman Rank we have converted them to a Cramér's V: value which is normalized between 0 and 1. This was done to make the values in the matrix more uniform however we must note that Cramér's V and Spearman's correlation coefficients are fundamentally different statistics and generally can't be directly compared.

Corelation With the Target Variable

Our target variable TARGET show whether the given application had any late payments (value = 1), we can see that no single feature is strongly correlated with it:

	Coefficient	P-Value
Column
ExtSource3	-0.161	0.000
ExtSource1	-0.131	0.000
ExtSource2	-0.128	0.000
MeanbureaudaysCredit	0.093	0.000
OccupationType	0.075	0.000
DaysEmployed	0.074	0.000
PrevRatioRejectedAccepted	0.073	0.000
PrevRatioRejectedAccepted_cats_2	0.072	0.000
PrevRatioRejectedAccepted_cats	0.072	0.000
OrganizationType	0.069	0.000
NameEducationType	0.067	0.000
PrevAmtDownPaymentSum	-0.057	0.000
PrevCreditReceivedRequestedDiff	0.055	0.000
DaysBirth	0.053	0.000
PrevLastLoanGoodsCategory	0.051	0.000
OwnCarAge	0.050	0.000
MeanbureauamtCreditSumDebt	0.049	0.000
MeanbureauamtCreditMaxOverdue	0.044	0.000
DaysIdPublish	0.042	0.000
CodeGender	0.041	0.000
PrevAvgYieldGroup	0.040	0.000
FlagDocument3	0.039	0.000
AmtGoodsPrice	-0.034	0.000
MaxbureaudaysCreditEnddate	0.034	0.000
NameFamilyStatus	0.027	0.002
AmtCredit	-0.023	0.001
AmtAnnuity	0.003	0.664

The chart below shows the relationship between selected categorical variables and loan status. E.g. a significantly higher proportion of loans taken out by males had issues.

CategoricalDtype(categories=['< 25% Rejected', '> 25% Rejected', 'All Accepted', 'No Previous App.'], ordered=False, categories_dtype=object)

Relationships Between Numerical and Categorical Variables

The charts below show pairs of numerical and categorical features (including some binned numerical features) that have a signficant relationships and at least a small effect size (eta_squared>0.01) based on the non-parametric Kruskal-Wallis Test (one-way ANOVA on ranks) testing whether samples originate from the same distribution.

*It's similar to the Mann–Whitney U test but allows comparing more than 2 groups

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

V:\projects\ppuodz-ML.4.1\shared\graph.py:1477: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = _df.groupby(c)[y_target]
V:\projects\ppuodz-ML.4.1\shared\graph.py:1490: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  group_counts = _df.groupby(c).size()

Analyzing Credit Scores (ExtSource1)

ExtSource1/2/3 are the variables most strongly correlated with the target variable, they indicate client credit scores obtained from external sources. While th correlation coeficients are very low (only slightly above 0.1) we'll look a bit more into these scores because ussually credit ratings tend be the most useful metric when estimating the risk of specific loans:

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\2151574185.py:16: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed two minor releases later. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap(obj)`` instead.
  colors = plt.cm.get_cmap('tab10', 4)

Summary for combined model:
                           Logit Regression Results                           
==============================================================================
Dep. Variable:                 TARGET   No. Observations:               109589
Model:                          Logit   Df Residuals:                   109585
Method:                           MLE   Df Model:                            3
Date:                Mon, 29 Apr 2024   Pseudo R-squ.:                  0.1047
Time:                        19:57:19   Log-Likelihood:                -25636.
converged:                       True   LL-Null:                       -28634.
Covariance Type:            nonrobust   LLR p-value:                     0.000
==============================================================================
                 coef    std err          z      P>|z|      [0.025      0.975]
------------------------------------------------------------------------------
const          0.6002      0.040     14.829      0.000       0.521       0.680
ExtSource1    -2.0989      0.061    -34.382      0.000      -2.219      -1.979
ExtSource2    -1.9640      0.060    -32.654      0.000      -2.082      -1.846
ExtSource3    -2.7793      0.062    -44.483      0.000      -2.902      -2.657
==============================================================================

This is a simple Logistic model that just uses the credit scores to estimate the target variable. The confidence interval shows the the standard deviation of the residuals from a combined logistic regression model (residuals in this context are the differences between the observed values (y_combined) and the predicted probabilities).

Gennerally the explained variabity (Pseudo R-squared) is very quite low at only 0.1047 however the model itself is statistically significant (LLR p-value = 0.0)

	Coefficient	Standard Error	P-Value	Conf. Interval Lower	Conf. Interval Upper
const	0.600	0.040	0.0	0.521	0.680
ExtSource1	-2.099	0.061	0.0	-2.219	-1.979
ExtSource2	-1.964	0.060	0.0	-2.082	-1.846
ExtSource3	-2.779	0.062	0.0	-2.902	-2.657

Normalized credit ratings from three sources are inversely related to default risk, with ExtSource3 having the strongest influence. We can see that a basic Logistic model can already provide a reasonably high result (AUC = 0.74). However, we have to note that the results are based on the full training set and are only provided for EDA/feature analysis purposes. Full statistical modelling will be done in further sections.

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:6: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:8: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 0][col], label=f'{col} - No Default', shade=True)

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:6: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:8: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 0][col], label=f'{col} - No Default', shade=True)

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:6: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\1233466688.py:8: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['TARGET'] == 0][col], label=f'{col} - No Default', shade=True)

We can see that while the the external credit are clearly related to default risk their explanatory power is somewhat limited because there is still a large amount of overlap (especially for ExtSource2, however it's coeefficient in our logistical model is similar to that of ExtSource1.

Previous Application History

Did any clients had previously applied for loans with Home Credit and what were the outcomes of their applications?

PrevRatioRejectedAccepted_cats
All Accepted        190370
> 25% Rejected       66215
< 25% Rejected       34079
No Previous App.     16847
Name: count, dtype: int64

Did any applicants default on any previous loans?

TotalDefaults_cats
No Defaults          304114
1 Defaulted Loans      3397
Name: count, dtype: int64

Suprisingly we can see that a ~1% of all applicants who were granted a loans have previously had payment difficulties with a previous loans at Home Credit. This is quite interesting considering that gennerally credit instituions are reluctant to offer loans again to problematic clients.

Total "Defaults"/Loans With Payment Difficulties per applicant:

	TotalDefaults	count	proportion
0	0.0	304114	0.99
1	1.0	3177	0.01
2	2.0	163	0.00
3	3.0	38	0.00
4	4.0	11	0.00
5	5.0	4	0.00
6	6.0	3	0.00
7	7.0	1	0.00

Previous Loan History and Default Risk

The chart below shows the default rate based on whether applicant has previous applied for loans with Home Cred: No Previous App. - no previous applications for client found (i.e. new clients) All Accepted - all previous applications were accepted < 25% Rejected - less than 1/4 applications were rejected > 25% Rejected - more than 1/4 applications were rejected

Interestingly we can see that while applicants whose previous loans were rejected are significantly more likely to default when finally given a loan previous clients with no failed applications have a higher default risk than new clients.

This likely limits the usefulness of the previous_application table because only a small proportion of clients have previously rejected applications

<Figure size 1000x600 with 0 Axes>

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\4178975199.py:5: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousRejections'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\4178975199.py:7: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousRejections'] == 0][col], label=f'{col} - No Rejections', shade=True)

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:10: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:12: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 0][col], label=f'{col} - No Rejections', shade=True)

<Figure size 1200x600 with 0 Axes>

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:10: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:12: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 0][col], label=f'{col} - No Rejections', shade=True)

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:10: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 1][col],
C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\3528939979.py:12: FutureWarning: 

`shade` is now deprecated in favor of `fill`; setting `fill=True`.
This will become an error in seaborn v0.14.0; please update your code.

  sns.kdeplot(data=features_matrix[features_matrix['AnyPreviousDefaults'] == 0][col], label=f'{col} - No Rejections', shade=True)

We can clearly see that clients who had run into payment issues with their past loans tend to have a signficantly lower credit ExtSource3 however there is almost no difference with other scores. This incidates that the data from Home Credit itself is only included in the third rating (which might explain its higher explantatory power in our Logistic model)

AmtIncomeTotal
AmtCredit
AmtAnnuity
AmtGoodsPrice
AmtReqCreditBureauHour
AmtReqCreditBureauDay
AmtReqCreditBureauWeek
AmtReqCreditBureauMon
AmtReqCreditBureauQrt
AmtReqCreditBureauYear
MaxbureauamtAnnuity
MaxbureauamtCreditMaxOverdue
MaxbureauamtCreditSum
MaxbureauamtCreditSumDebt
MaxbureauamtCreditSumLimit
MaxbureauamtCreditSumOverdue
MeanbureauamtAnnuity
MeanbureauamtCreditMaxOverdue
MeanbureauamtCreditSum
MeanbureauamtCreditSumDebt
MeanbureauamtCreditSumLimit
MeanbureauamtCreditSumOverdue
MinbureauamtAnnuity
MinbureauamtCreditMaxOverdue
MinbureauamtCreditSum
MinbureauamtCreditSumDebt
MinbureauamtCreditSumLimit
MinbureauamtCreditSumOverdue
PrevAmtApplicationMean
PrevAmtApplicationSum
PrevAmtCreditMean
PrevAmtCreditSum
PrevAmtDownPaymentSum

Loan Purposes

V:\projects\ppuodz-ML.4.1\shared\graph.py:1529: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  

C:\Users\Paulius\AppData\Local\Temp\ipykernel_29624\4000396085.py:4: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  prop_df = features_matrix.groupby('NameContractType')['TARGET'].value_counts(normalize=True).unstack().fillna(0)

<Figure size 1000x600 with 0 Axes>

EDA Summary

The EDA was performed in paralel with performing feature enginerring (aggregation of non-main tables) and building an initial LGBM model (using all features), to minimize unnecessary complexity only features which have some importance { > X } are included.

index SK_BUREAU_ID not found in dataframe, creating new integer column

Appending previous history

Missing Values

'AMT_CREDIT'

	TARGET	NameContractType	CodeGender	FlagOwnCar	FlagOwnRealty	CntChildren	AmtIncomeTotal	AmtCredit	AmtAnnuity	AmtGoodsPrice	NameTypeSuite	NameIncomeType	NameEducationType	NameFamilyStatus	NameHousingType	RegionPopulationRelative	DaysBirth	DaysEmployed	DaysRegistration	DaysIdPublish	OwnCarAge	FlagMobil	FlagEmpPhone	FlagWorkPhone	FlagContMobile	FlagPhone	FlagEmail	OccupationType	CntFamMembers	RegionRatingClient	RegionRatingClientWCity	WeekdayApprProcessStart	HourApprProcessStart	RegRegionNotLiveRegion	RegRegionNotWorkRegion	LiveRegionNotWorkRegion	RegCityNotLiveCity	RegCityNotWorkCity	LiveCityNotWorkCity	OrganizationType	ExtSource1	ExtSource2	ExtSource3	ApartmentsAvg	BasementareaAvg	YearsBeginexpluatationAvg	YearsBuildAvg	CommonareaAvg	ElevatorsAvg	EntrancesAvg	FloorsmaxAvg	FloorsminAvg	LandareaAvg	LivingapartmentsAvg	LivingareaAvg	NonlivingapartmentsAvg	NonlivingareaAvg	ApartmentsMode	BasementareaMode	YearsBeginexpluatationMode	YearsBuildMode	CommonareaMode	ElevatorsMode	EntrancesMode	FloorsmaxMode	FloorsminMode	LandareaMode	LivingapartmentsMode	LivingareaMode	NonlivingapartmentsMode	NonlivingareaMode	ApartmentsMedi	BasementareaMedi	YearsBeginexpluatationMedi	YearsBuildMedi	CommonareaMedi	ElevatorsMedi	EntrancesMedi	FloorsmaxMedi	FloorsminMedi	LandareaMedi	LivingapartmentsMedi	LivingareaMedi	NonlivingapartmentsMedi	NonlivingareaMedi	FondkapremontMode	HousetypeMode	TotalareaMode	WallsmaterialMode	EmergencystateMode	Obs30CntSocialCircle	Def30CntSocialCircle	Obs60CntSocialCircle	Def60CntSocialCircle	DaysLastPhoneChange	FlagDocument2	FlagDocument3	FlagDocument4	FlagDocument5	FlagDocument6	...	MinbureauamtCreditSum	MinbureauamtCreditSumDebt	MinbureauamtCreditSumLimit	MinbureauamtCreditSumOverdue	MinbureaucntCreditProlong	MinbureaucreditDayOverdue	MinbureaudaysCredit	MinbureaudaysCreditEnddate	MinbureaudaysCreditUpdate	MinbureaudaysEnddateFact	MinbureauskIdBureau	ActiveLoansCount	TotalDefaults	DefaultRatio	LastLoanIssuedDays	PrevAmtApplicationMean	PrevAmtApplicationSum	PrevAmtCreditMean	PrevAmtCreditSum	PrevAmtDownPaymentSum	PrevRateInterestPrimaryMean	PrevRateInterestPrimaryStd	PrevTotalDpdSum	PrevHasAnyDpdMean	PrevHasAnyDpdSum	PrevMonthsWithDpdPropMean	PrevMonthsWithDpdPropSum	PrevTotalPreviousLoans	PrevCreditReceivedRequestedDiff	PrevRatioSumDownPaymentCredit	PrevLastLoanInterestRate	PrevLastLoanPurpose	PrevLastLoanContractStatus	PrevLastLoanDecisionDate	PrevLastLoanPaymentType	PrevLastLoanCodeRejectReason	PrevLastLoanClientType	PrevLastLoanPortfolio	PrevLastLoanGoodsCategory	PrevLastLoanProductType	PrevLastLoanYieldGroup	PrevContractStatusApprovedCount	PrevContractStatusCanceledCount	PrevContractStatusRefusedCount	PrevContractStatusUnusedofferCount	PrevPortfolioCardsCount	PrevPortfolioCarsCount	PrevPortfolioCashCount	PrevPortfolioPosCount	PrevPortfolioXnaCount	PrevProductTypeXnaCount	PrevProductTypeWalkinCount	PrevProductTypeXsellCount	PrevAvgYieldGroup	PrevDaysAfterFirstApplication	PrevCurrentlyActiveLoans	PrevApprovedLoans	PrevCanceledLoans	PrevRefusedLoans	PrevUnusedofferLoans	PrevTotalLoans	PrevAcceptedToTotalRatio	PrevCancelledToTotalRatio	PrevRefusedToTotalRatio	PrevUnusedToTotalRatio	PrevRatioRejectedAccepted	PrevCodeRejectReasonClientCount	PrevCodeRejectReasonHcCount	PrevCodeRejectReasonLimitCount	PrevCodeRejectReasonScoCount	PrevCodeRejectReasonScofrCount	PrevCodeRejectReasonSystemCount	PrevCodeRejectReasonVerifCount	PrevCodeRejectReasonXapCount	PrevCodeRejectReasonXnaCount	PrevLastLoanNflagInsuredOnApproval	PrevAvgNflagInsuredOnApproval	ExtSource2_binned	ExtSource3_binned	ExtSource1_binned	DaysEmployed_binned	AmtCredit_binned	OwnCarAge_binned	PrevCreditReceivedRequestedDiff_binned	DaysBirth_binned	PrevAmtDownPaymentSum_binned	MeanbureauamtCreditSumDebt_binned	MaxbureaudaysCreditEnddate_binned	PrevAvgYieldGroup_binned	MeanbureauamtCreditMaxOverdue_binned	AmtGoodsPrice_binned	DaysLastPhoneChange_binned	PrevRatioRejectedAccepted_binned	DaysIdPublish_binned	AmtAnnuity_binned	AmtIncomeTotal_binned	FlagDocument3_binned	MaxbureaudaysCredit_binned	PrevAcceptedToTotalRatio_binned	MaxbureaudaysEnddateFact_binned
0	0	Cash loans	M	True	False	2	207000.0	465457.5	52641.0	418500.0	Unaccompanied	Commercial associate	Secondary / secondary special	Married	House / apartment	0.009630	-13297	-762	-637.0	-4307	19	1	1	0	1	0	0	Sales staff	4	2	2	THURSDAY	11	0	0	0	0	1	1	Business Entity Type 3	0.675878	0.604894	0.000527	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	0	0	0	0	-2.0	0	1	0	0	0	...	19449.0	0.00	0.0	0.0	0.0	0.0	-2857.0	-2710.0	-1033.0	-1036.0	5022129.0	3.0	0.0	0.0	-225	22279.500000	111397.50	100030.500000	500152.50	6642.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	8.0	388755.0	0.013280	NaN	XNA	Refused	-1291.0	Cash through the bank	HC	Repeater	Cash	XNA	walk-in	high	2.0	0.0	3.0	0.0	2.0	0.0	1.0	2.0	0.0	2.0	3.0	0.0	1.666667	-546.0	NaN	2.0	0.0	3.0	0.0	5.0	0.400000	0.000000	0.600000	0.000000	1.000000	0.0	3.0	0.0	0.0	0.0	0.0	0.0	2.0	0.0	NaN	0.000000	0.57 - 0.66	0.0 - 0.37	0.68 - 0.96	-1211.0 - -289.0	270000.0 - 512338.5	15 - 91	117146.25 - 10985769.0	-15739.0 - -12412.0	6111.0 - 17820.0	0.0 - 43926.19	120.0 - 906.0	1.33 - 2.0	0.0 - 2076.99	238500.0 - 450000.0	-279.0 - 0.0	0.25 - 13.6	-6551.0 - -4305.0	34587.0 - 230161.5	202500.0 - 117000000.0	1	-143.0 - -1.0	0.0 - 0.5	-347.0 - -153.0
1	0	Cash loans	F	True	True	0	247500.0	1281712.5	48946.5	1179000.0	Unaccompanied	Commercial associate	Higher education	Single / not married	House / apartment	0.006852	-14778	-1141	-1610.0	-4546	11	1	1	0	1	0	1	Managers	1	3	3	THURSDAY	10	0	0	0	0	0	0	Business Entity Type 3	0.430827	0.425351	0.712155	0.0753	0.0568	0.9970	0.9592	0.1326	0.08	0.0517	0.4167	0.2917	0.0735	0.0601	0.0844	0.0058	0.1118	0.0756	0.0566	0.9940	0.9216	0.0523	0.0806	0.0345	0.3333	0.0417	0.0445	0.0652	0.0857	0.0	0.0000	0.0760	0.0568	0.9970	0.9597	0.1335	0.08	0.0517	0.4167	0.2917	0.0748	0.0611	0.0859	0.0058	0.1142	reg oper account	block of flats	0.0754	Monolithic	False	2	0	2	0	-1071.0	0	1	0	0	0	...	190867.5	0.00	0.0	0.0	0.0	0.0	-1487.0	-391.0	-765.0	-765.0	5977750.0	2.0	0.0	0.0	-52	454259.250000	5451111.00	488020.875000	5856250.50	36603.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	8.0	405139.5	0.006250	NaN	XAP	Refused	-757.0	XNA	HC	Repeater	Cards	XNA	walk-in	XNA	10.0	0.0	2.0	0.0	2.0	0.0	7.0	3.0	0.0	3.0	3.0	6.0	1.000000	-245.0	NaN	10.0	0.0	2.0	0.0	12.0	0.833333	0.000000	0.166667	0.000000	0.181818	0.0	1.0	1.0	0.0	0.0	0.0	0.0	10.0	0.0	NaN	0.555556	0.39 - 0.57	0.67 - 0.89	0.33 - 0.51	-1211.0 - -289.0	808650.0 - 4050000.0	9 - 15	117146.25 - 10985769.0	-15739.0 - -12412.0	17820.0 - 3960000.0	-220213.42 - 0.0	1682.0 - 31199.0	0.5 - 1.0	2076.99 - 47406123.0	679500.0 - 4050000.0	-1571.0 - -762.0	0.0 - 0.25	-6551.0 - -4305.0	34587.0 - 230161.5	202500.0 - 117000000.0	1	-300.0 - -143.0	0.8 - 1.0	-153.0 - 0.0
2	0	Cash loans	F	True	False	0	202500.0	495000.0	39109.5	495000.0	Unaccompanied	Working	Secondary / secondary special	Married	House / apartment	0.035792	-17907	-639	-2507.0	-1461	4	1	1	1	1	0	0	Sales staff	2	2	2	TUESDAY	16	0	0	0	0	0	0	Self-employed	0.527239	0.531760	0.207964	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	5	0	5	0	-1435.0	0	1	0	0	0	...	179550.0	0.00	0.0	0.0	0.0	0.0	-2305.0	-2123.0	-1706.0	-2123.0	5353341.0	3.0	0.0	0.0	-394	121017.818571	847124.73	126556.547143	885895.83	12330.9	NaN	NaN	0.0	0.0	0.0	0.0	0.0	13.0	38771.1	0.013919	NaN	XNA	Canceled	-107.0	XNA	XAP	Repeater	XNA	XNA	XNA	XNA	3.0	2.0	1.0	1.0	0.0	0.0	1.0	2.0	4.0	6.0	0.0	1.0	1.000000	-2835.0	NaN	3.0	2.0	1.0	1.0	7.0	0.428571	0.285714	0.142857	0.142857	0.250000	1.0	1.0	0.0	0.0	0.0	0.0	0.0	5.0	0.0	NaN	0.000000	0.39 - 0.57	0.0 - 0.37	0.51 - 0.68	-1211.0 - -289.0	270000.0 - 512338.5	0 - 5	16020.0 - 117146.25	-19681.0 - -15739.0	6111.0 - 17820.0	141482.43 - 43650000.0	906.0 - 1682.0	0.5 - 1.0	NaN	450000.0 - 679500.0	-1571.0 - -762.0	0.0 - 0.25	-1728.0 - 0.0	34587.0 - 230161.5	148500.0 - 202500.0	1	-624.0 - -300.0	0.0 - 0.5	-718.0 - -347.0
3	0	Cash loans	F	False	True	0	247500.0	254700.0	24939.0	225000.0	Unaccompanied	State servant	Secondary / secondary special	Widow	House / apartment	0.046220	-19626	-6982	-11167.0	-3158	<NA>	1	1	0	1	0	0	High skill tech staff	1	1	1	FRIDAY	14	0	0	0	0	0	0	Business Entity Type 3	NaN	0.693521	0.614414	0.1320	0.0645	0.9846	NaN	NaN	0.16	0.0690	0.6250	NaN	NaN	NaN	0.1628	NaN	0.0022	0.1345	0.0670	0.9846	NaN	NaN	0.1611	0.0690	0.6250	NaN	NaN	NaN	0.1696	NaN	0.0023	0.1332	0.0645	0.9846	NaN	NaN	0.16	0.0690	0.6250	NaN	NaN	NaN	0.1657	NaN	0.0022	NaN	NaN	0.1285	Panel	False	0	0	0	0	-2000.0	0	1	0	0	0	...	38268.0	0.00	0.0	0.0	0.0	0.0	-2657.0	-2322.0	-2321.0	-2321.0	5347375.0	0.0	0.0	0.0	-1234	104292.000000	104292.00	103153.500000	103153.50	10431.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	7.0	-1138.5	0.101121	NaN	XAP	Approved	-2000.0	Cash through the bank	XAP	New	POS	Audio/Video	XNA	high	1.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	2.000000	-2599.0	NaN	1.0	0.0	0.0	0.0	1.0	1.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	0.0	0.000000	0.66 - 0.85	0.54 - 0.67	NaN	-17912.0 - -2760.25	45000.0 - 270000.0	NaN	-3960000.0 - -558.0	-19681.0 - -15739.0	6111.0 - 17820.0	-220213.42 - 0.0	-41875.0 - 120.0	1.33 - 2.0	2076.99 - 47406123.0	45000.0 - 238500.0	-4131.0 - -1571.0	0.0 - 0.25	-3269.0 - -1728.0	24916.5 - 34587.0	202500.0 - 117000000.0	1	-2922.0 - -624.0	0.8 - 1.0	-2858.0 - -718.0
4	0	Cash loans	M	False	True	0	112500.0	308133.0	15862.5	234000.0	Unaccompanied	Working	Secondary / secondary special	Single / not married	House / apartment	0.018850	-20327	-1105	-7299.0	-494	<NA>	1	1	0	1	0	0	Laborers	1	2	2	WEDNESDAY	11	0	0	0	0	0	0	Business Entity Type 3	0.654882	0.560690	0.636376	0.0619	0.0553	0.9717	NaN	NaN	0.00	0.1724	0.1667	NaN	0.0866	NaN	0.0749	NaN	0.0149	0.0630	0.0574	0.9717	NaN	NaN	0.0000	0.1724	0.1667	NaN	0.0885	NaN	0.0780	NaN	0.0158	0.0625	0.0553	0.9717	NaN	NaN	0.00	0.1724	0.1667	NaN	0.0881	NaN	0.0762	NaN	0.0152	NaN	block of flats	0.0765	Stone, brick	False	0	0	0	0	-173.0	0	1	0	0	0	...	28575.0	0.00	0.0	0.0	0.0	0.0	-2618.0	-2430.0	-2415.0	-2430.0	6293239.0	1.0	0.0	0.0	-693	70180.000000	631620.00	177702.500000	1599322.50	2475.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	7.0	967702.5	0.001548	NaN	XNA	Canceled	-173.0	XNA	XAP	Repeater	XNA	XNA	XNA	XNA	5.0	1.0	3.0	0.0	2.0	0.0	4.0	2.0	1.0	3.0	1.0	5.0	1.666667	-2504.0	NaN	5.0	1.0	3.0	0.0	9.0	0.555556	0.111111	0.333333	0.000000	0.500000	0.0	2.0	0.0	1.0	0.0	0.0	0.0	6.0	0.0	NaN	0.600000	0.39 - 0.57	0.54 - 0.67	0.51 - 0.68	-1211.0 - -289.0	270000.0 - 512338.5	NaN	117146.25 - 10985769.0	-25201.0 - -19681.0	0.0 - 6111.0	141482.43 - 43650000.0	906.0 - 1682.0	1.33 - 2.0	NaN	45000.0 - 238500.0	-279.0 - 0.0	0.25 - 13.6	-1728.0 - 0.0	1980.0 - 16573.5	25650.0 - 112500.0	1	-300.0 - -143.0	0.5 - 0.8	-718.0 - -347.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
99995	0	Revolving loans	F	False	True	1	202500.0	585000.0	29250.0	585000.0	Unaccompanied	Working	Secondary / secondary special	Married	House / apartment	0.010147	-13827	-1317	-398.0	-1172	<NA>	1	1	0	1	0	0	Sales staff	3	2	2	MONDAY	9	0	0	0	0	0	0	Self-employed	0.678014	0.591704	0.456110	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	0	0	0	0	-967.0	0	0	0	0	0	...	0.0	0.00	0.0	0.0	0.0	0.0	-1043.0	-766.0	-710.0	-766.0	6624105.0	5.0	0.0	0.0	-504	38610.000000	115830.00	38770.500000	116311.50	9220.5	NaN	NaN	0.0	0.0	0.0	0.0	0.0	7.0	481.5	0.079274	NaN	XNA	Approved	-371.0	XNA	XAP	Repeater	Cash	XNA	x-sell	middle	3.0	0.0	0.0	0.0	0.0	0.0	1.0	2.0	0.0	2.0	0.0	1.0	1.000000	-2887.0	NaN	3.0	0.0	0.0	0.0	3.0	1.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	3.0	0.0	1.0	0.333333	0.57 - 0.66	0.37 - 0.54	0.68 - 0.96	-2760.25 - -1211.0	512338.5 - 808650.0	NaN	-558.0 - 16020.0	-15739.0 - -12412.0	6111.0 - 17820.0	0.0 - 43926.19	906.0 - 1682.0	0.5 - 1.0	0.0 - 2076.99	450000.0 - 679500.0	-1571.0 - -762.0	0.0 - 0.25	-1728.0 - 0.0	24916.5 - 34587.0	148500.0 - 202500.0	0	-624.0 - -300.0	0.8 - 1.0	-718.0 - -347.0
99996	0	Cash loans	M	False	True	0	225000.0	562500.0	31540.5	562500.0	Unaccompanied	Working	Secondary / secondary special	Married	House / apartment	0.018209	-20956	-3053	-13427.0	-4280	<NA>	1	1	0	1	1	0	Drivers	2	3	3	FRIDAY	12	0	0	0	0	0	0	Business Entity Type 3	NaN	0.140261	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	2	1	2	1	-842.0	0	1	0	0	0	...	3127500.0	NaN	NaN	0.0	0.0	0.0	-1914.0	994.0	-498.0	-498.0	5160093.0	0.0	0.0	0.0	-498	375000.000000	1125000.00	531804.000000	1595412.00	0.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	7.0	470412.0	0.000000	NaN	XAP	Approved	-542.0	XNA	XAP	Repeater	Cards	XNA	x-sell	XNA	3.0	0.0	0.0	0.0	1.0	0.0	2.0	0.0	0.0	0.0	1.0	2.0	1.500000	-2842.0	NaN	3.0	0.0	0.0	0.0	3.0	1.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	3.0	0.0	0.0	0.333333	0.0 - 0.39	NaN	NaN	-17912.0 - -2760.25	512338.5 - 808650.0	NaN	117146.25 - 10985769.0	-25201.0 - -19681.0	0.0 - 6111.0	NaN	906.0 - 1682.0	1.33 - 2.0	NaN	450000.0 - 679500.0	-1571.0 - -762.0	0.0 - 0.25	-4305.0 - -3269.0	24916.5 - 34587.0	202500.0 - 117000000.0	1	-2922.0 - -624.0	0.8 - 1.0	-718.0 - -347.0
99997	0	Revolving loans	M	True	False	1	135000.0	180000.0	9000.0	180000.0	Family	Working	Higher education	Married	House / apartment	0.035792	-10578	-592	-5307.0	-3257	11	1	1	0	1	0	0	Core staff	3	2	2	SATURDAY	14	0	0	0	0	0	0	Self-employed	0.602777	0.487365	0.490258	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	0	0	0	0	-2521.0	0	0	0	0	0	...	91593.0	49016.43	0.0	0.0	0.0	0.0	-229.0	206.0	-3.0	NaN	5967472.0	2.0	0.0	0.0	0	103455.000000	103455.00	93109.500000	93109.50	10345.5	NaN	NaN	0.0	0.0	0.0	0.0	0.0	10.0	-10345.5	0.111111	NaN	XAP	Approved	-320.0	Cash through the bank	XAP	New	POS	Computers	XNA	middle	1.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	1.000000	-2702.0	2.0	1.0	0.0	0.0	0.0	1.0	1.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	0.0	0.000000	0.39 - 0.57	0.37 - 0.54	0.51 - 0.68	-1211.0 - -289.0	45000.0 - 270000.0	0 - 5	-3960000.0 - -558.0	-12412.0 - -7673.0	6111.0 - 17820.0	141482.43 - 43650000.0	906.0 - 1682.0	0.5 - 1.0	0.0 - 2076.99	45000.0 - 238500.0	-4131.0 - -1571.0	0.0 - 0.25	-3269.0 - -1728.0	1980.0 - 16573.5	112500.0 - 148500.0	0	-143.0 - -1.0	0.8 - 1.0	NaN
99998	0	Cash loans	F	True	True	0	135000.0	254700.0	17149.5	225000.0	Unaccompanied	Working	Incomplete higher	Single / not married	Rented apartment	0.002506	-8062	-92	-6446.0	-724	13	1	1	0	1	0	0	High skill tech staff	1	2	2	TUESDAY	7	0	0	0	1	1	0	Housing	0.352214	0.714284	0.651260	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	<NA>	1	0	1	0	-707.0	0	1	0	0	0	...	540000.0	0.00	0.0	0.0	0.0	0.0	-473.0	-106.0	-100.0	-106.0	5414423.0	0.0	0.0	0.0	-106	396369.000000	396369.00	380272.500000	380272.50	39640.5	NaN	NaN	0.0	0.0	0.0	0.0	0.0	10.0	-16096.5	0.104242	NaN	XAP	Approved	-707.0	Cash through the bank	XAP	New	POS	Audio/Video	XNA	low_normal	1.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	0.000000	-1406.0	NaN	1.0	0.0	0.0	0.0	1.0	1.000000	0.000000	0.000000	0.000000	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	0.0	0.000000	0.66 - 0.85	0.54 - 0.67	0.33 - 0.51	-289.0 - 365243.0	45000.0 - 270000.0	9 - 15	-3960000.0 - -558.0	-12412.0 - -7673.0	17820.0 - 3960000.0	-220213.42 - 0.0	-41875.0 - 120.0	0.0 - 0.5	0.0 - 2076.99	45000.0 - 238500.0	-762.0 - -279.0	0.0 - 0.25	-1728.0 - 0.0	16573.5 - 24916.5	112500.0 - 148500.0	1	-624.0 - -300.0	0.8 - 1.0	-153.0 - 0.0
99999	0	Cash loans	M	True	True	2	157500.0	746280.0	59094.0	675000.0	Unaccompanied	Working	Higher education	Married	House / apartment	0.020246	-13934	-1210	-3758.0	-5221	19	1	1	0	1	0	1	Drivers	4	3	3	WEDNESDAY	11	0	0	0	0	0	0	Self-employed	0.409607	0.366777	0.502878	0.0062	NaN	0.9886	NaN	NaN	NaN	0.2069	0.0417	NaN	NaN	NaN	NaN	NaN	NaN	0.0063	NaN	0.9886	NaN	NaN	NaN	0.2069	0.0417	NaN	NaN	NaN	NaN	NaN	NaN	0.0062	NaN	0.9886	NaN	NaN	NaN	0.2069	0.0417	NaN	NaN	NaN	NaN	NaN	NaN	NaN	block of flats	0.0162	Stone, brick	False	0	0	0	0	-2452.0	0	1	0	0	0	...	37800.0	0.00	0.0	0.0	0.0	0.0	-2708.0	-1073.0	-892.0	-1073.0	6671057.0	1.0	0.0	0.0	-552	48712.500000	194850.00	48766.500000	195066.00	4725.0	NaN	NaN	0.0	0.0	0.0	0.0	0.0	6.0	216.0	0.024223	NaN	XAP	Approved	-174.0	XNA	XAP	Repeater	POS	Auto Accessories	XNA	middle	3.0	0.0	1.0	0.0	0.0	0.0	0.0	4.0	0.0	4.0	0.0	0.0	1.250000	-2237.0	NaN	3.0	0.0	1.0	0.0	4.0	0.750000	0.000000	0.250000	0.000000	0.250000	0.0	0.0	0.0	1.0	0.0	0.0	0.0	3.0	0.0	0.0	0.000000	0.0 - 0.39	0.37 - 0.54	0.33 - 0.51	-1211.0 - -289.0	512338.5 - 808650.0	15 - 91	-558.0 - 16020.0	-15739.0 - -12412.0	0.0 - 6111.0	43926.19 - 141482.43	120.0 - 906.0	1.0 - 1.33	NaN	450000.0 - 679500.0	-4131.0 - -1571.0	0.0 - 0.25	-6551.0 - -4305.0	34587.0 - 230161.5	148500.0 - 202500.0	1	-2922.0 - -624.0	0.5 - 0.8	-718.0 - -347.0

100000 rows × 250 columns

Tight layout not applied. tight_layout cannot make axes width small enough to accommodate all axes decorations

set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.

set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipykernel_17429/466980156.py in ?()
      1 importlib.reload(graph)
      2 
----> 3 graph.summary_df_features(features_matrix_sorted_imp)

~/data/projects/ppuodz-ML.4.1/shared/graph.py in ?(source_df)
   1207             plt.show()
   1208 
   1209         except Exception as ex:
   1210             plt.close(fig)
-> 1211             raise ex

~/data/projects/ppuodz-ML.4.1/shared/graph.py in ?(source_df)
   1207             plt.show()
   1208 
   1209         except Exception as ex:
   1210             plt.close(fig)
-> 1211             raise ex

~/miniconda3/envs/rapids_v2/lib/python3.10/site-packages/scipy/stats/_stats_py.py in ?(a, axis, nan_policy)
   1988     if contains_nan and nan_policy == 'omit':
   1989         a = ma.masked_invalid(a)
   1990         return mstats_basic.normaltest(a, axis)
   1991 
-> 1992     s, _ = skewtest(a, axis)
   1993     k, _ = kurtosistest(a, axis)
   1994     k2 = s*s + k*k
   1995 

~/miniconda3/envs/rapids_v2/lib/python3.10/site-packages/scipy/stats/_stats_py.py in ?(a, axis, nan_policy, alternative)
   1602 
   1603     if axis is None:
   1604         a = np.ravel(a)
   1605         axis = 0
-> 1606     b2 = skew(a, axis)
   1607     n = a.shape[axis]
   1608     if n < 8:
   1609         raise ValueError(

~/miniconda3/envs/rapids_v2/lib/python3.10/site-packages/scipy/stats/_axis_nan_policy.py in ?(***failed resolving arguments***)
    519                 # behavior of those would break backward compatibility.
    520 
    521                 if sentinel:
    522                     samples = _remove_sentinel(samples, paired, sentinel)
--> 523                 res = hypotest_fun_out(*samples, **kwds)
    524                 res = result_to_tuple(res)
    525                 res = _add_reduced_axes(res, reduced_axes, keepdims)
    526                 return tuple_to_result(*res)

~/miniconda3/envs/rapids_v2/lib/python3.10/site-packages/scipy/stats/_stats_py.py in ?(a, axis, bias, nan_policy)
   1190         a = ma.masked_invalid(a)
   1191         return mstats_basic.skew(a, axis, bias)
   1192 
   1193     mean = a.mean(axis, keepdims=True)
-> 1194     m2 = _moment(a, 2, axis, mean=mean)
   1195     m3 = _moment(a, 3, axis, mean=mean)
   1196     with np.errstate(all='ignore'):
   1197         zero = (m2 <= (np.finfo(m2.dtype).resolution * mean.squeeze(axis))**2)

~/miniconda3/envs/rapids_v2/lib/python3.10/site-packages/scipy/stats/_stats_py.py in ?(a, moment, axis, mean)
   1065             n_list.append(current_n)
   1066 
   1067         # Starting point for exponentiation by squares
   1068         mean = (a.mean(axis, keepdims=True) if mean is None
-> 1069                 else np.asarray(mean, dtype=dtype)[()])
   1070         a_zero_mean = a - mean
   1071 
   1072         eps = np.finfo(a_zero_mean.dtype).resolution * 10

TypeError: float() argument must be a string or a real number, not 'NAType'

Clustering

Clustering before using XGBoost can simplify data and possibly improve model performance by highlighting patterns that XGBoost may overlook. This preprocessing step reduces dimensionality and can enhance model interpretability, but its effectiveness depends on data relevance and feature importance evaluation.

K-Prototypes

Most suitable the dataset has clear boundaries and a roughly uniform distribution for optimal results. We've been unable to obtain clearly defined cluster when using it and based on the type of the dataset it's probably not the most suitable algorithm.

DBSCAN

Is an unsupervised algorithm which is more suitable for datasets with significant noise or irrelevant data points (e.g. data exhibits non-globular or irregularly shaped clusters)

Feature

• We've experimented

https://www.kaggle.com/c/home-credit-default-risk/data

AutoML - EvalML

Additionally, we've included a model selected using EvalML (an auto ML library) and a raw dataset (with Featuretools aggregations etc.) [TODO: include notebook]